Normative Substance Use Antecedents among Sexual Minorities: A Scoping Review and Synthesis

Abstract

Although sexual minority stress remains the dominant perspective for understanding disproportionate substance use among lesbian, gay and bisexual (LGB) populations, social norms are among the most predictive and commonly targeted substance use antecedents in other high-risk groups. This scoping review seeks to bring clarity to the body of norms-focused alcohol, tobacco, and other drug (ATOD) research conducted with LGBs over the past 20 years, identify intervention implications, and present priority directions for future research. Fifty-two peer-reviewed studies published between June 1999 and June 2019 were identified from searches of PubMed, PsycInfo, and Medline databases using combinations of terms related to: social norms; sexual orientation or sexual minority status; and, the use of alcohol, tobacco, and other drugs. A far greater number of studies focused on actual ATOD norms than perceived ATOD norms or discrepancies between actual or perceived norms, illuminating the need for additional research focused on these levels of analysis. Taken together, this literature suggests that: (1) perceived ATOD norms are reliable predictors of LGBs’ ATOD use; (2) actual ATOD use norms are low among LGBs participating in representative, population-based survey studies; and, (3) LGBs over-estimate the ATOD use of peers. Thus, intervention strategies including personalized normative feedback, psychoeducation, and social branding may hold promise in reducing LGBs substance use. However, additional research is needed to increase our understanding of injunctive ATOD norms, identify meaningful LGB reference groups, elucidate environmental influences on ATOD norms, and examine relationships between stigma experiences, perceived norms, and ATOD use.

Research over the past 20 years has reliably found that lesbian and bisexual women are more likely to engage in hazardous drinking and experience alcohol use disorders than are heterosexual women, while gay and bisexual men are more likely to use and become dependent on illicit drugs than heterosexual men (see Green & Feinstein, 2012 for a review). Further, gay men, lesbian women, and bisexual men are more likely than heterosexuals of their same age and sex to smoke cigarettes (Balsam, Beadnell, & Riggs, 2012; Hoffman, Delahanty, Johnson, & Zhao, 2018; Watson, Lewis, Fish, & Goodenow, 2018) and use other tobacco products (Fallin-Bennett, Lisha, & Ling, 2017). Although targeted evidence-based interventions developed to prevent or reduce use of alcohol, tobacco, and other drugs (ATODs) among sexual minorities remain scarce (Blume, 2016; Cochran & Mays, 2016), much recent research has sought to inform intervention development by identifying factors that account for sexual minorities’ disproportionate substance use. The majority of this work has been informed by Meyer’s (2003) sexual minority stress framework, which posits that LGB individuals’ higher rates of substance use are consequences of possessing a stigmatized, socially disadvantaged sexual minority identity. In support of Meyer’s model, extant research has established links between internalized sexual minority stigma, prejudice, discrimination, and coping-motivated ATOD use in LGB populations (Feinstein & Newcomb, 2016; Goldbach, Tanner-Smith, Bagwell, & Dunlap, 2014; Lehavot & Simoni, 2011; Livingston, Christianson, & Cochran, 2016; Pachankis, Hatzenbuehler, & Starks, 2014; Slater, Godette, Huang, Ruan, & Kerridge, 2017; Wilson, Gilmore, Rhew, Hodge, & Kaysen, 2016). These findings have encouraged the development of programs aiming to reduce substance use by increasing LGB individuals’ abilities to respond to society’s stigma with healthier, adaptive coping strategies (Chaudoir, Wang, & Pachankis, 2017; Pachankis, Hatzenbuehler, Rendina, Safren, & Parsons, 2015; Pachankis et al., 2014). Although sexual minority stress undoubtedly plays a role in the observed ATOD disparities, and these interventions appropriately follow from theory, this literature has been criticized for inattention to developmental and socio-cultural factors that may also contribute to increased substance use among sexual minorities (Frost, 2017; Hatzenbuehler, 2009; Hatzenbuehler, Corbin, & Fromme, 2008; Hunt, Antin, Sanders, & Sisneros, 2019; Mereish, Goldbach, Burgess, & DiBello, 2017).

The past two decades have produced a large literature revealing perceived social norms to be among the most predictive and commonly targeted substance use antecedents in other populations burdened by disproportionate use including college students, homeless youth, active duty military personnel, and veterans (Barman-Adhikari, Craddock, Bowen, Das, & Rice, 2018; Larimer & Cronce, 2007; Pemberton et al., 2011; Walters & Neighbors, 2005). However, norms focused ATOD research with sexual minority populations has been slower to emerge, and researchers have described the literature as in an infancy stage of development (Mereish, 2018; Talley et al., 2016). As such, this scoping review seeks to characterize the ways in which substance use norms have been defined and operationalized in sexual minority research conducted during this period, synthesize findings, and identify critical gaps in the literature that could inform the development of norms-based interventions for sexual minority populations.

Theories of Normative Influence and Operationalizations of Norms

Social psychological theories of normative influence and subsequent ATOD norms research with other populations have operationalized normative constructs at three levels of analysis: perceived norms, actual norms, and discrepancies between perceived and actual norms. These levels of analysis provide a valuable organizational framework for this review and thus theoretical background, key definitions, and methods are introduced for each.

Perceived Norms.

The focus theory of normative conduct (Cialdini, 2012; Cialdini, Kallgren, & Reno, 1991; Kallgren, Reno, & Cialdini, 2000), the theory of normative social behavior (Lapinski & Rimal, 2005; Rimal, 2008; Rimal & Real, 2005), and social norms theory (Berkowitz, 2004; Perkins & Berkowitz, 1986; Perkins, 2003) similarly assert that perceptions of behavioral and attitudinal norms in self-relevant peer groups are among the strongest predictors of future behavior. Perceived descriptive norms refer to a person’s perceptions of the behaviors engaged in by members of their peer group, while perceived injunctive norms refer to a person’s perceptions of the behaviors approved of, or deemed acceptable, by peers. Extant research has documented strong prospective relationships between of perceptions of substance use among members of the peer-group and subsequent substance use among sexual majority adolescents (Oluwoye, Merianos, & Nabors, 2017), college students (Larimer & Cronce, 2007; Walters & Neighbors, 2005), homeless youth (Barman-Adhikari et al., 2018), working adults (Hester, Squires, & Delaney, 2005), and military personnel (Pemberton et al., 2011). In research with these populations, surveys have prompted participants to first estimate the substance use approval and substance use behaviors of a “typical” member of the peer group and then report their own substance use behaviors and attitudes. This work has revealed two variables closely related to the strength of the perceived norm-behavior relationship: specificity and proximity of the peers referenced. Reference group specificity refers to the amount of demographic information (e.g., age, sex, race, sexual identity) provided about the “typical peer” in question (Borsari & Carey, 2003; Larimer et al., 2009), while psychological proximity is a related construct indicative of how interpersonally close (e.g., close friend, family member) or distant (e.g., unspecified member of a social group) the target peer is to the participant (LaBrie, Hummer, & Lac, 2011). In general, higher reference group specificity and proximity increase the degree to which the peers in question are similar and meaningful to the perceiver, thereby strengthening the perceived norms-behavior relationship (LaBrie et al., 2011; Larimer et al., 2009; Liu, Zhao, Chen, Falk, & Albarracín, 2017). In addition, much social psychological research has established that for less proximal groups, the strength of the relationship between perceived norms and behavior is dependent on the degree to which the reference group is meaningful and important to the perceiver (e.g., Lapinski & Rimal, 2005; Rimal, 2008; Terry, Hogg, & White, 1999).

Actual Norms.

Social norms theory (Perkins & Berkowitz, 1986) and the theory of normative social behavior (Lapinski & Rimal, 2005; Rimal, 2008) distinguish between perceived norms and “actual” or “collective” norms. These latter norms describe a group’s prevailing reality or culture and are most often assessed by querying group members about their attitudes, behaviors, and experiences individually, aggregating responses, and then deriving either a statistical average or rate of the behavior, attitude, or experience in the group. Researchers aggregating individual responses at the group level to identify an average or majority response also commonly use terms such as “norm”, “normative”, and “normalization” to describe average reports of substance use approval by group members, as well as the contexts and motivations for substance use reported by the numerical majority (e.g., Dumbili & Williams, 2017; Voigt, 2015; Weiss & Woods, 2018; Yang & Bissell, 2017).

Discrepancies between Actual and Perceived Norms.

Social norms theory (SNT; Berkowitz, 2004; Perkins et al., 1986; Perkins, 2003) is uniquely focused on the likelihood of discordance between actual and perceived norms. SNT explains that individuals tend to overestimate the degree to which peers approve of, and engage in, substance use and observable public behavior is an influential source of information that fuels misperceptions of peer ATOD norms (Berkowitz, 2004; Perkins, 2003). That is, over-estimated norms tend to result from observing a minority of the peer group publicly engage in drinking, smoking, or drug use. These substance use observations are more salient and memorable than more responsible and mundane behaviors that may be more common in the group but are less visible. Problematically, among non and light substance users, much research suggests that over-estimating the ATOD use and approval of self-relevant peers increases the likelihood that the individual initiates or escalates their substance use in order to approximate the normative standard they misperceive for the group. Further, among heavy substance users, over-estimations of peers’ substance use and approval act as a barrier to behavior change by functioning to maintain an individual’s perception that their behavior is aligned with the group norm and thus, not problematic or unusual (Berkowitz, 2004; Miller & Prentice, 2016; Perkins, 2003). Discrepancies between actual and perceived norms are identified by comparing perceived “typical peer” substance use norms with corresponding actual norms for the group derived either from the same convenience sample or an appropriate representative, population-based secondary data source (Miller & Prentice, 2016).

Established Norms-based Intervention Strategies

In university and military populations found to over-estimate peer ATOD norms, misperceptions have been corrected, and substance use thereby reduced, by brief personalized normative feedback (PNF) interventions (Miller & Prentice, 2016). These interventions simply require members of a social group to answer survey questions about their perceptions of the “typical” group member’s substance use and then report on their own substance use. Behavioral averages for the group are computed and participants are presented with individualized reports which graphically highlight how much and how often they thought others in their group use the substance, how much and how often other group members actually use the substance, and how their own substance use compares to actual substance use norms of the self-relevant group (Miller & Prentice, 2016; Perkins et al., 1986). To date, brief PNF interventions have most frequently targeted drinking on college campuses and have been efficacious in reducing students’ frequency of quantity of alcohol use (Borsari & Carey, 2003; Dotson, Dunn, & Bowers, 2015; Miller & Prentice, 2016). However, drug-specific PNF has also effectively motivated tobacco dependent adults to seek treatment (Steinberg, Ziedonis, Krejci, & Brandon, 2004), decreased negative consequences among marijuana users (Palfai et al., 2014), reduced use and markers of dependence among non-treatment seeking MDMA users (Martin & Copeland, 2010), and decreased rates of relapse and recidivism among multi-drug offenders (Yokotani & Tamura, 2015).

Whereas PNF interventions are designed to reveal and reinforce low substance use norms when group members’ perceptions of norms are inflated relative to reality, psychoeducational and social branding campaigns may be more appropriate when harmful ATOD norms are accurately perceived by group members. For a specific substance, psychoeducation might consist of peer-led educational activities in community spaces designed to change substance-related knowledge, risk perceptions, and decrease permissive attitudes toward use, while increasing awareness about local recovery and treatment resources (e.g., Bennett, Patterson, Reynolds, Wiitala, & Lehman, 2004; Perry & Jessor, 1985; Power & Hunter, 2001). Although psychoeducational components target permissive ATOD injunctive norms and increase awareness of treatment and recovery resources, they may also include social marketing and branding elements which leverage social identity, normative influence, commercial marketing strategies in order to change peer norms around a target substance or health behavior (Diamond et al., 2009; Hoek & Jones, 2011; Ling, Holmes, Jordan, Lisha, & Bibbins-Domingo, 2017).

The Current Review

As the past 20 years have reflected a golden age of norms-focused theory, research and subsequent substance use intervention development for sexual majority populations, this scoping review focuses on this period and seeks to answer the research question, “What has empirical research conducted with sexual minorities found with regard to substance use norms?” As this research question is broad and quite nuanced given the variety of different ways social norms have been operationalized, and there is a clear need to identify themes and gaps in this body of literature, we elected to conduct a scoping review rather than a more rigorous, methodologically focused systematic review (Armstrong, Hall, Doyle, & Waters, 2011).

Method

Our methodology followed the PRISMA guidelines for scoping reviews (Tricco et al., 2018). Figure 1 illustrates the literature search and article identification process. Peer-reviewed research studies published between June 1, 1999 and June 1, 2019 were identified from PubMed, PsycINFO, and Medline database titles and abstract searches using combinations of terms related to (1) social norms and social factors (2) sexual orientation, sexual minority identity or behavior; and, (3) the use of alcohol, tobacco, and other drugs. See Appendix A for all search term combinations and PubMed-specific syntax. The 3 database searches yielded a total of 181 unique references. The abstracts of these references were screened by two authors and 98 articles were excluded on the basis being in a language other than English (n=3), not discussing alcohol, tobacco, or other drug use (n=33), not discussing norms, culture, or social factors (n=42), or not including sexual minorities in the sample (n=20). This left a total of 83 articles for full-manuscript review. Upon review of these manuscripts, 31 additional papers were excluded due to not discussing or assessing ATOD specific norms/culture/or social factors (n=21) or not including analysis between sexual orientation subgroups or within a sexual minority sub-group (n=3). The remaining 52 studies met inclusion criteria and are included in this scoping review.

Figure 1.

Literature search and article screening process.

To characterize the scope of this literature and elucidate gaps in knowledge, studies were first classified in terms of their operationalization of normative constructs. Then, years of publication, substance use outcomes, sample, and design characteristics were charted overall and by their primary operationalization of norm-related constructs. Representative methods and findings at each level of analysis are reviewed and themes identified.

Results

The 52 studies meeting inclusion criteria were published in journals spanning clinical psychology, social psychology, community psychology, sociology, public health, and behavioral medicine. Most studies (85%; n=44) focused on the psychological identity dimension of sexual orientation and either examined ATOD norms within a sample of LGB participants or compared ATOD norms between mixed LGB and heterosexually identified participants. See Table 1 for a numerical breakdown of sample, design, and methodological characteristics overall and within each level of analysis.

Table 1.

A numerical breakdown of the sample, method, and design characteristics of the 52 articles meeting inclusion criteria by level of analysis and overall

	Perceived Norms (n=18)	Actual Norms (n=29)	Discrepancies (n=5)	Overall (n=52)
	# of studies	# of studies	# of studies	# of studies
Publication Year
1999–2003	1	3	-	4
2004–2008	3	10	-	13
2009–2013	6	7	1	14
2014–2019	8	9	4	21
Sample
Representative/population-based	6	19	-	25
Convenience	12	10	5	27
Domain of Sexual Orientation
Sexual identity	16	24	4	44
Sexual behavior	2	5	1	8
Age-Group
Youth/adolescents	3	2	-	5
Emerging adults/college students	3	4	-	7
Adults	12	23	5	40
Method & Design
Cross-sectional survey	14	20	2	36
Longitudinal survey	4	2	1	7
Qualitative study	-	5	1	6
Intervention evaluation study	-	2	1	3
Type of Norms Assessed
Descriptive only	3	21	5	29
Injunctive only	1	3	-	4
Descriptive & injunctive	3	2	-	5
Norms combined with other constructs	1	1	-	2
Neither type of norm directly assessed	10	2	-	12
Substances Studied
Alcohol	6	7	2	15
Tobacco	2	5	-	7
Other drugs	2	6	2	10
Alcohol & other drugs	4	7	1	12
Alcohol, tobacco, & other drugs	4	4	-	8

Perceived Norms

Only 7 of the 52 studies meeting inclusion criteria (13%) directly examined perceived ATOD norms in relation to substance use. At the lowest level of reference group specificity and proximity, Hatzenbuehler, Corbin, and Fromm (2008) followed a large, mixed sexual orientation sample of adolescents as they transitioned from high school to college. Surveys prompted participants to estimate the alcohol consumption (descriptive norm) and drinking approval (i.e., injunctive norm) of the average male and female peer without specification of the referent peer’s sexual identity. Perceptions of peer drinking and approval for drinking were found to be positive predictors of both sexual minority and majority participants’ future drinking. However, relative to heterosexuals, LGB participants perceived the average peer to consume more alcohol, be more approving of drinking, and also reported consuming greater quantities of alcohol themselves. Further, sexual orientation-based differences in alcohol consumption were partially mediated by perceptions of injunctive and descriptive peer drinking norms. That is, although question prompts did not specify the sexual identity of the referent peer, sexual minority participants reported higher perceptions of average peer drinking and approval for drinking than did heterosexual participants, which, in turn, partially explained the relationship between sexual orientation and alcohol use in the sample. A recent cross-sectional study built on these findings by increasing psychological proximity of referent peers to “close friends” and extending the alcohol focus to tobacco, marijuana, and prescription drugs (Mereish et al., 2017). Consistent with alcohol findings from Hatzenbuehler and colleagues’ (2008), LGB adolescents reported greater lifetime and recent use of the four substances as well as elevated perceptions of close friends’ substance use, relative to heterosexual adolescents. Further, the composite measure of perceived substance use norms fully accounted for sexual orientation-based differences in ATOD use.

Whereas the measures of perceived norms in these studies were marked by low specificity, not distinguishing between perceptions of norms for sexual minority versus heterosexual peers, research with adults suggests the importance of normative perceptions specific to peers sharing the participants’ own sex and sexual minority status. For example, a prospective survey study investigating relationships between perceived descriptive drinking norms and alcohol use among heterosexual, lesbian, and bisexual women found that sexual minority women consistently perceived the “typical sexual minority woman” to consume more alcohol than “the typical woman” across 4 assessment points (Litt, Lewis, Rhew, Hodge, & Kaysen, 2015). Further, sexual minority-specific descriptive peer norms were stronger positive predictors of lesbian and bisexual women’s future drinking than were less specific norms for the “typical woman.” In another recent study, Ehlke, Stamates, Kelley, & Braitman (2019) recruited a sample of bisexual women and separately assessed participants’ perceived drinking norms for bisexual, heterosexual, lesbian, and bisexual women. Underscoring the importance of reference group specificity and sexual minority social identity, perceived bisexual drinking norms were most strongly related to bisexual participants’ own alcohol consumption while perceived heterosexual drinking norms were negatively associated with participants’ drinking.

Three additional studies deviated from traditional questions about “typical peer” substance use and approval to simply ask participants about the degree to which friends and others close to them use substances and approve of use (Hamilton & Mahalik, 2009; Lea et al., 2019; Tobin, Davey-Rothwell, Yang, Siconolfi, & Latkin, 2014), thereby increasing the psychological proximity of referent peers while decreasing specificity. For example, Tobin and colleagues (2014) queried an online sample of African American men who have sex with men about the alcohol use and approval of their friends and found that perceptions of these friend-specific norms were positively related to the participants’ own alcohol consumption and high-risk drinking. Similarly, Hamilton and Mihalik (2009) cross-sectionally examined gay men’s perceptions of descriptive norms for health-risk behaviors including use of alcohol, tobacco, illicit drugs, and risky sexual practices and assessed perceived norms for 4 reference groups varying in psychological proximity: (a) male friends, (b) male relatives, (c) male coworkers, and (d) gay men. The researchers examined the degree to which these perceived norms, masculine gender norms, internalized homophobia, and experiences with prejudice and discrimination accounted for participants’ use of alcohol, tobacco, illicit drugs, and risky sexual practices. Results revealed the composite measure of perceived norms to be most strongly related to participants’ own risk-behaviors. However, analyses did not disaggregate reference groups to identify the level of proximity or specificity at which perceived norms were most strongly related to substance use.

A larger proportion of studies meeting inclusion criteria (19%, n=10) did not directly assess perceived substance use norms, but rather linked social and environmental factors to increased ATOD use and theorized that perceptions of LGB specific peer norms may explain these relationships. For instance, several survey studies with gay men have revealed an elevated likelihood of drug use among participants who reside in gay enclaves, have networks composed predominantly of other gay men, and have greater socialization with gay men (Buttram & Kurtz, 2013; Carpiano, Kelly, Easterbrook, & Parsons, 2011; Green, 2003; Kurtz, 2009). Similarly, among LGB youth and emerging adults, greater participation in the LGB community, and LGB nightlife in particular, have also been cross-sectionally (Baiocco, D’Alessio, & Laghi, 2010; Demant, Hides, White, & Kavanagh, 2018; Stalgaitis, Wagner, Navarro, Hoffman, & Delahanty, 2018) and longitudinally (Rosario, Schrimshaw, & Hunter, 2004) linked to increased alcohol, marijuana, and tobacco use. Likewise, one of the few studies to investigate substance use in a large sample of LGB older adults (50+ years or older) revealed an unexpected positive relationship between LGB specific social support and high-risk drinking among lesbian and bisexual older women (Bryan, Kim, & Fredriksen-Goldsen, 2017). The authors speculate that social support may be likely to come from other sexual minority women and occur in bars where alcohol use and heavy drinking are highly visible, thereby elevating perceived drinking norms. Thus, researchers sampling gay men, older lesbian and bisexual women, and LGB adolescents have similarly theorized that elevated perceptions of peer substance use norms are one unmeasured explanatory mechanism by which LGB environmental and social network factors relate to substance use.

Actual Norms

More than half of the studies meeting inclusion criteria (56%; n=29) focused on actual substance use norms specific to prevalence, approval, contexts or motivations for use. Eight epidemiological studies with population-based samples of adults examined actual descriptive substance use and dependence norms by sexual orientation finding increased odds of use and dependence among sexual minority respondents. Specifically, sexual minoritity women have been found more likely to drink alcohol, engage in heavy drinking, and become dependent on alcohol than their sexual majority female peers, regardless of whether sexual orientation has been defined in terms of sexual identity or behavior (Cochran, Ackerman, Mays, & Ross, 2004; McCabe, Hughes, Bostwick, West, & Boyd, 2009; Wilsnack et al., 2008). In contrast, sexual minority men have rarely been found to differ from sexual majority men in their alcohol consumption, but data suggest that they have a significantly greater likelihood of using and becoming dependent on drugs including cocaine, heroin, and methamphetamine (Cochran et al., 2004; McCabe et al., 2009; Schuler, Rice, Evans-Polce, & Collins, 2018). Data also suggest that sexual minority men and women are significantly more likely to smoke and use tobacco products than their sexual majority counterparts (e.g., Balsam et al., 2012; Fallin-Bennett et al., 2017; Hoffman et al., 2018; Watson et al., 2018).

Most relevant to our analysis of actual descriptive substance use norms is McCabe and colleagues (2009) emphasis on the overall low base rates of hazardous drinking, smoking, illicit drug use, and substance dependence observed in sexual minority populations despite well-documented sexual orientation-based ATOD disparities. For instance, although the prevalence of substance use and dependence were significantly higher among sexual minority than majority respondents in their nationally representative sample of adults, the highest weighted prevalence of past-year heavy drinking and alcohol dependence among sexual minority women were respectively 25% and 16%. Similarly, among sexual minority men, the highest illicit drug use prevalence was 18% with a drug dependence prevalence of 5%. Thus, although members of sexual minority groups reliably report greater substance use and dependence than sexual majority group members, in no representative, population-based study to date have heavy drinking, tobacco use, illicit drug use, or dependence on any substance found to be normative in the sense of being reported by a majority of group members.

Several studies (Cochran, Grella, & Mays, 2012; Jasinski & Ford, 2007; Hinds, Loukas, & Perry, 2019) sought to investigate whether sexual identity-based differences in adults’ and adolescents’ ratings of substance use approval and acceptability (i.e., actual injunctive norms; explained sexual identity-based differences in substance use (i.e. actual descriptive norms). For instance, Cochran and colleagues (2012) used data from a telephone survey of Californians to examine potential differences in substance approval and availability between sexual minority and sexual majority participants. Relative to heterosexuals, LGBs reported greater drug availability, more accepting attitudes toward illicit drug use and heavy drinking, and greater consumption of alcohol and drugs. Further, substance availability and attitudes toward heavy drinking and drug use mediated differences in past year heavy drinking and illicit drug use reported by sexual minority and majority participants. Thus, in this study differences in substance use by LGB and heterosexual participants (actual descriptive norms) were partially explained by sexual-identity based differences in accepting substance-related attitudes (actual injunctive norms).

On the surface, these findings imply that approval for substance use is a normative attitude among LGBs accounting for normative substance use in this population. However, a closer look at descriptive statistics reveal that neither approving attitudes nor substance use were “normative” in the sense of being reported by a majority of LGB participants in either study. For example, in Cochran, Grella, & Mays (2012) study, although the degree of substance use disapproval was stronger among heterosexual participants, the average LGB respondent still strongly disapproved of using illicit drugs including cocaine and hallucinogens and somewhat disapproved of using marijuana and drinking heavily. Echoing emphasis by McCabe et al. (2009), the base rates of heavy drinking and illicit drug use among LGBs in this representative sample of Californians were also relatively low (22% reported past year marijuana use, 19% reported past year other drug use, and less than 10% reported weekly heavy drinking) providing additional support for the notion that substance use and dependence are concentrated among a minority of LGBs.

A study with a representative sample of U.S. adults also highlighted sexual identity-based differences in actual norms for visiting heavy drinking and drug use contexts including bars, dance clubs, and circuit parties (Trocki, Drabble, & Midanik, 2005). Relative to 15% of heterosexual women, roughly 30% of lesbians and 33% of bisexual women reported visiting bars on a monthly or more frequent basis. Among men, the pattern differed slightly with only 22% of heterosexual and 29% of bisexual men men reported visting bars at least monthly, while 55% of gay men reported this behavior. Several studies also established links between more frequent LGB bar and club attendance and heavier substance use (Baiocco et al., 2010; Kipke et al., 2007). Indeed, relative to the small proportions of LGBs who report substance use and dependence in representative, population-based studies, when convenience samples of LGB bar, club, and party attendees have been surveyed about their substance use on site, much larger proportions report ATOD use (Toby Lea et al., 2019; Lea, Reynolds, & de Wit, 2013). In light of these findings, subsequent work has sought to better understand motivations for alcohol and drug use in LGB community nightlife settings. For instance, several qualitative studies with lesbian-identified adults suggest that lesbian bar patronage is a central aspect of the lesbian lifestyle reflecting two distinct normative motivations for alcohol use (Cogger, Conover, & Israel, 2012; Drabble & Trocki, 2014; Gruskin, Byrne, Kools, & Altschuler, 2006; Parks, 1999; Trocki et al., 2005). Participants emphasized the importance of lesbian bars to both sexual identity development and socialization, describing alcohol as a social lubricant aiding the formation of relationships with other lesbians. In addition to these social motivations for drinking, participants also understood the centrality of the lesbian bar and presence of alcohol use to be related to the prejudice historically experienced by lesbians, with many women interviewed also describing stress-coping motivations for drinking.

Studies examining normative motivations for drug use among gay and bisexual men reveal a more complicated picture wherein different substances are used for different reasons. Some substances (i.e., poppers, methamphetamine, GHB, ketamine) are used by sexual minority men to enhance sexual encounters and decrease sexual inhibition, a practice known as “chemsex” (Bourne et al., 2015). However, research suggests that only a minority of sexual minority men take part in chemsex and other drugs including MDMA, marijuana, and cocaine are more commonly used in gay bars, nightclubs, and circuit parties to forge connections with other gay and bisexual men, foster feelings of gay community belonging, and facilitate a sense of a shared experience (Bauermeister, 2007; Halkitis & Palamar, 2008; Kecojevic, Corliss, & Lankenau, 2015; Lea et al., 2016; Ross, Mattison, & Franklin, 2003). Other research suggests that prescription drugs, as well as those serving sexual and social functions, are also commonly used by gay and bisexual men to cope with stress, stigma, and negative emotions (Bauermeister, 2007; Kecojevic et al., 2015; Lea et al., 2016). In sum, it appears that illicit drugs are primarily used by gay and bisexual men for many of the same social and coping reasons that alcohol is used by lesbian women.

In addition, two community-based intervention studies sought to augment actual norms around tobacco use and distance smoking from LGB culture via psychoeducation and social branding campaigns targeting LGB community spaces (Fallin, Neilands, Jordan, & Ling, 2015; Plant et al., 2017). For example, the Los Angeles Department of Public Health’s Break Up With Tobacco campaign (Plant et al., 2017) included both psychoeducation and social marketing components to target LGB nightlife venues in Los Angeles County. The campaign included bold, colorful advertisements that featured clever statements an LGB smoker might say when ending their relationship with cigarettes, a quit-line phone number, and the address of the campaign website on which visitors could access tobacco harm information and cessation resources. Simultaneously, a scantily clad LGB peer outreach team called the Break Up Squad held multiple events at in community spaces to facilitate LGB peer-to-peer conversations about the negative effects of tobacco use. Follow-up data revealed that campaign exposure predicted individuals’ contemplating cessation and taking initial steps to quit. However, similar to the social branding campaign study by Fallin et al. (2015), changes in norms were not directly assessed and study designs did not feature control groups.

Discrepancies between Actual and Perceived Norms

Only 10% (n=5) of the studies meeting inclusion criteria investigated potential discrepancies between perceived and actual substance use norms among sexual minorities, and notably all of these studies were published quite recently (since 2013). First, in a prospective study, Ramchand and colleagues (2013) queried gay and bisexual male participants attending a weekend dance party about their perceptions of drug use (e.g., cocaine, ecstasy, cocaine/crack, poppers, etc.) among gay/bisexual males attending the same party and assessed participants’ own drug use intentions. Participants over-estimated the use of ecstasy, cocaine/crack, non-prescribed erectile dysfunction drugs, and poppers among gay and bisexual male peers attending the party and these misperceptions of gay and bisexual peer’s drug use predicted use assessed at following the party among participants both intending and not intending to use drugs at the beginning of the weekend.

Shifting focus to alcohol use among lesbian-identified women, Boyle, LaBrie, & Witkovic (2016) surveyed an online sample of California-residing lesbians about their perceptions of “the typical California lesbian’s” quantity and frequency of alcohol use and their own alcohol use. The researchers compared estimates of lesbian peer drinking to actual alcohol use in the sample as well as representative lesbian drinking data from the California Health Interview Survey (CHIS), finding that participants over-estimated lesbian peers’ quantity and frequency of alcohol consumption according to actual drinking norms in the sample and population-based estimates from the CHIS. Further, after controlling for participants own alcohol consumption, larger discrepancies between actual and perceived drinking norms were significantly associated with more frequent lesbian bar patronage and use of popular social media sites (e.g., Facebook, Instagram) suggesting that observations of peers in these physical and virtual spaces may inflate perceived lesbian peer drinking norms.

Moreover, a qualitative study investigating perceptions of chemsex found that English gay men perceived most of their gay male peers to engage in this practice (Ahmed et al., 2016). Meanwhile, representative public health data suggested that only 12% of the gay male population in England engages in chemsex, confirming that participants’ perceptions were wildly exaggerated. Further, when asked to identify sources of information that contributed to their chemsex perceptions, participants emphasized that observations of gay male peers both on social/hook-up apps and in nightclubs and contributed to their estimations.

While the other discrepancy-focused studies queried LGB community members about their perceptions of quantity and frequency substance use norms, Boyle and colleagues (2017) examined the accuracy of normative perceptions specific to the stigma-coping motivated alcohol and other drug use of LGB peers. Surveying LGB Americans following the Pulse nightclub shooting, a shared sexual minority stress event, revealed that the majority of participants over-estimated the proportion of same-sexual identity peers who used alcohol and other drugs to cope following the Pulse tragedy. Among the 10 coping behaviors assessed, LGB peers were perceived as most likely to have engaged in: posting their feelings on social media (82%); seeking social support (75%); using alcohol (69%); and, using other drugs (41%) to cope with the Pulse tragedy. Although large proportions of the sample reported themselves posting their feelings on social media (75%) and seeking social support (71%), significantly smaller proportions reported using alcohol (26%) or other drugs (7%) to cope with this event.

Only 1 study meeting inclusion criteria explored the utility of a personalized normative feedback substance-specific intervention. Kuerbis and colleagues (2014) provided a small group of problem drinking men (n=101) who reported having sex with men their percentile rank in terms of quantity and frequency of drinking compared to adult males in the US, and examined discrepancies between perceptions and actual norms. The majority of participants over-estimated US male drinking norms and responded positively to normative feedback. Further, significant reductions in alcohol consumption were reduced between baseline and follow-up assessments.

Discussion

Although additional research is clearly needed at perceived norm and discrepancy levels of analysis, taken together the studies reviewed suggest that: (1) perceived ATOD norms are reliable predictors of LGBs’ ATOD use; (2) actual ATOD use norms are low among LGBs participating in representative, population-based survey studies; however, these norms are higher when convenience samples are recruited from LGB nightlife settings; (3) LGBs over-estimate the ATOD use of peers with misperceptions potentially related to observations of peers in LGB nightlife settings and on social media sites and dating/hook-up apps. Collectively, this small body of research suggests that substance use norms in LGB populations’ function very similarly to those in university and military populations also known to disproportionately use substances. Given these similarities, it follows that the personalized normative feedback interventions and psychoeducational/social branding campaigns that have been effective in reducing substance use among members of these groups may also have utility in LGB communities. However, out of the 52 studies included in this review, only 5% (n=3) described or evaluated intervention strategies and problematically, none of these studies featured control groups. Given the low actual LGB substance use and approval norms revealed in population-based surveys, initial findings suggesting that LGB community members over-estimate these norms, and the effectiveness of PNF interventions in other populations, PNF represents an understudied but promising potential strategy for reducing ATOD use among sexual minorities. However, as the smallest proportion of studies included in this review focused on norms at the discrepancy level of analysis (10%; n=5), additional research investigating discrepancies between perceived and actual substance use norms is needed to ensure the appropriateness of this approach for understudied substances (e.g., tobacco, cannabis) and sexual minority sub-groups (e.g., bisexuals, youth).

A general trend observed among the studies reviewed at the perceived and actual norm levels of analysis was that intervention implications associated with findings were either not articulated or rarely went beyond a broad statement about the potential utility of norms-based interventions (for valuable exceptions see Ehlke et al., 2019 and Litt et al., 2015). To better inform future research and intervention development, a second, increasingly focused generation of norms-focused research with sexual minorities should ensure that operationalizations of norms, assessment methods, reference groups, and discussions of findings more clearly map onto a specific level of analysis and follow from social psychological theories of normative influence. Further, discussions of study findings should aim for greater specificity in connecting findings back to theory and delineating specific, established norms-based strategies that may have utility in the context of findings. The sections that follow highlight additional gaps in the literature gleaned from this review and identify several critical research questions to be addressed by the next generation of norms-focused ATOD research.

Priority Directions for Future Research

Investigate Injunctive ATOD Norms.

Across the 3 levels of analyses, a far greater number of studies directly assessed descriptive (n=29) than injunctive (n=4) ATOD norms and further, several studies combined measures of descriptive and injunctive norms in analysis (n=5). This is problematic in the context of the theory of normative social behavior (Lapinski & Rimal, 2005; Rimal, 2008; Rimal & Real, 2005) which identifies injunctive norms as unique predictors of behavior and explains that descriptive and injunctive norms are not always consistent. That is, it is possible for a person to believe that use of a substance is common among group members (a perceived descriptive norm), and simultaneously believe that the group as a whole frowns upon regular or heavy use (a perceived injunctive norm). In this situation, the TNSB predicts that perceived injunctive norms will moderate the relationship between perceived descriptive norms and the perceiver’s substance use. Thus, one priority direction for future research with sexual minorities is to examine perceived injunctive norms as separate predictors of future substance use as well as potential moderators of relationships between perceived descriptive norms and future substance use. It may be particularly valuable to assess injunctive ATOD norms among LGBs frequenting bars, clubs, and circuit parties. Although ATOD use is common and highly observable in these settings, LGB community members may still privately disapprove of heavy or frequent use, condemn the concurrent use of multiple substances, or view some motivations for use as less acceptable than others. Assessing approval for high-risk substance use behaviors among LGB bar, club, and circuit party patrons (actual injunctive norms), along with perceptions of LGB peer approval for these behaviors (perceived injunctive norms) would allow researchers to determine if discrepancies exist and potentially inform the development of targeted norm-based interventions.

2. Examine perceived norms in meaningful, actionable reference groups.

Very few of the studies included in this review (n=4 at the perceived level; n=4 at the discrepancy level) employed “typical peer” norm items consistent with the larger norms literature; and further, there was great variability in the specificity and proximity of the reference groups from which norms were assessed. It is critical to emphasize that PNF, psychoeducation, and social branding strategies are similarly designed to target substance use norms in specific social groups and communities central to individuals’ sense of self. Thus, in some cases, the reference groups assessed in studies were not actionable from an intervention development standpoint. For example, PNF, social branding, and psychoeducation programs would have difficulty targeting substance use norms in higher proximity reference groups, for instance, among the sexual-identity non-specific close friends of individuals participating in a study (e.g., Mereish et al., 2017; Tobin et al., 2014). The efficacy of norms-based intervention strategies will also be diminished to the extent that less specific groups not relevant to the self are targeted. Although 4 of the 5 studies focused on the discrepancy of analysis followed the broader SNT literature in assessing perceptions of norms among referent peers high in specificity (sharing LGB identity with participants) but low in proximity (large social group rather than close friends), the only PNF intervention study identified problematically sampled treatment-seeking sexual minority men based on sexual behavior (i.e., MSM) then assessed perceived norms and delivered actual norms for the US male population as a whole (Kuerbis et al., 2014). To improve this literature substantially, future research across levels of analysis should seek to identify the sexual minority sub-groups and communities associated with the highest levels of social identification and salience of substance use norms. As a blueprint for future work, researchers might seek to improve upon the prospective studies conducted by Hamilton & Mahalik, (2009) and Litt et al. (2015) by assessing sexual minority participants’ “typical peer” descriptive and injunctive norms at several different levels of sexual minority reference group specificity (e.g., LGBTQ community, lesbian community, Los Angeles lesbian community) measuring level of social identification with each reference group, comparing the overall strengths of associations between reference group norms and substance use over time, and examining potential moderating effects of social identification.

3. Elucidate environmental influences on ATOD Norms.

Only 2 of the 52 studies included in this review examined social and environmental factors that might contribute to inflated perceptions of substance use norms among sexual minorities. As these studies point to bar patronage and social media/dating app use as variables that may influence perceptions of sexual minority-specific ATOD norms, additional research focused on these and other potential sources of normative influence are needed so that tailored norms-based intervention and prevention strategies can be developed to counter such influences. Relatedly, as many researchers have documented concerns about efforts by alcohol and tobacco companies to market their products at LGB bars, clubs, and pride events over the past 20 years (e.g., Dilley, Spigner, Boysun, Dent, & Pizacani, 2008; Drabble, 2000; Lee, Goldstein, Ranney, Crist, & McCullough, 2011; Spivey, Lee, & Smallwood, 2018), it was surprising that no studies identified by our method examined exposure to LGB-specific alcohol and tobacco marketing in relation to perceptions of alcohol and tobacco use among LGB peers. Thus, whether targeted LGB-focused alcohol and tobacco marketing and event sponsorship inflates perceptions of alcohol and tobacco use among LGB peers remains an important research question that may carry implications for marketing/sponsorship policies as well as norms-based alcohol and tobacco interventions. As exposure to marketing is inherently difficult to measure, it may be beneficial for future research to borrow and adapt the validated brand recognition tools (e.g., Braun et al., 2018; Meerkerk & van Straaten, 2019) and innovative methods (e.g., wearable cameras, fMRI, and geospatial positioning systems; Chambers et al., 2018; Courtney, Rapuano, Sargent, Heatherton, & Kelley, 2018) developed to study the impact of targeted alcohol and tobacco marketing in other populations.

4. Investigate relationships between stigma, perceived norms, and ATOD use.

Although much stigma-focused work informed by Meyer’s sexual minority stress model was published during the two decades targeted by this scoping review, notably only 3 of the 52 studies reviewed (6%; Boyle et al., 2017; Hamilton & Mahalik, 2009; Mereish et al., 2017) simultaneously assessed perceived substance use norms and sexual minority stigma-related variables as predictors of substance use. Thus, it appears that very little research has considered the possibility that norms and stigma are not competing explanations for ATOD use among sexual minorities, but rather complementary explanations which may independently predict and/or potentially interact to increase the likelihood of substance use beyond either factor alone. Despite the limited data investigating these possibilities, theories of intra-group dynamics suggest different ways in which experiences with sexual minority stigma might alter perceptions of norms, making misperceptions more likely, and/or strengthen the norms-behavior relationship. As theorized by Lick, Durso, and Johnson (2013), individuals impacted by sexual minority stigma may seek refuge in visible LGB community spaces and engage in substance use to cope with their stigmatizing experiences; other LGB individuals, unencumbered by stigma but motivated to socialize with peers and engage in LGB culture, may observe peers’ substance use in these settings and come to view these behaviors as normative, increasing the likelihood that they imbibe in the future. This theorized pattern is supported by studies suggesting that social and coping motivations play distinct roles in LGB substance use (e.g., Lea et al., 2016; Parks, 1999; Ross et al., 2003; Trocki et al., 2005). Additional support comes from the finding that perceptions of LGB peers’ coping motivated substance use following a stigmatizing event (i.e., the Pulse Nightclub Shooting) predicted participants own substance use over and above the stress impact of the event (Boyle et al., 2017).

Another compelling possibility is that sexual minority stigma experiences such as prejudice, harassment, and bullying, which originate from interactions with heterosexual out-group members, may strengthen the degree to which perceptions (and misperceptions) of LGB-specific substance norms contribute to the stigmatized individual’s substance use behaviors. This scenario is consistent with self-categorization theory (Hogg & Reid, 2006; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) which posits that when group memberships are made salient to an individual, as would be the case when an LGB individual is discriminated against, bullied, or harassed based on their sexual minority status, depersonalization and self-stereotyping processes are likely to occur. These processes, in turn, lead the individual to adopt the beliefs and behaviors they perceive to be normative within the group. An additional finding by Hamilton & Mahalik (2009) supports this pattern of relationships. Among the gay men surveyed, experiences with anti-gay prejudice moderated the relationship between their composite measure of perceived norms for several male reference groups and health-risk behaviors (including substance use) such that the relationship between perceived norms and behavior was stronger among those reporting more experience with anti-gay prejudice. Future research might seek to improve upon this cross-sectional study by prospectively assessing recent sexual minority stigma experiences, LGB specific perceived and actual norms for several substances, and social identification with an LGB community then examining whether experiences with sexual minority stigma strengthen prospective relationships between perceived ATOD norms and behavior. Replication of this pattern of results would also underscore the need for multi-component ATOD interventions that pair social-norms and sexual minority stress focused strategies, therefore representing a potentially fruitful direction for future research and theoretical development.

Limitations

This scoping review was limited to peer-reviewed articles published in English between 1999 and 2019 and did not include grey literature (e.g., unpublished manuscripts, dissertations, technical reports). We focused on research conducted during the 2 decades that directly followed major advances in norms-based social psychological theory, thus reflecting a period wherein comparatively large literatures examined norms as antecedents to substance use in sexual majority populations. Although the date range selected for this review reflect the golden age of norms-based substance use research, it is possible that ATOD norms were examined among sexual minority populations prior to this period. Further, review inclusion was restricted to studies that either assessed sexual orientation in terms of sexual identity or behavior, or explicitly focused on a sexual minority sample defined by these sexual orientation dimensions. Although no articles were excluded from this review due to assessing gender minority status in addition to sexual minority status, and umbrella terms including “queer”, “LGBT”, and “LGBTQ” were included search terms, gender minority specific terms alone were not included in our search. As this approach did not yield any norms-focused studies that identified or focused on gender minority participants, the extent to which norm-focused substance use research with gender minorities has been conducted remains unclear; and thus, represents yet another important question to be answered by future research.

Conclusion

Although social norms have only sporadically been studied as antecedents to substance use among sexual minorities over the past 20 years and this literature remains in its infancy, research suggests that perceived norms and discrepancies between perceived and actual norms appear to function as antecedents to substance use among sexual minorities in the same ways that they do in several high-risk sexual majority groups in which norms-based strategies have proven popular and effective. Given the dearth of targeted, evidence-based intervention and prevention strategies to reduce substance use in LGB communities, a focused second generation of norms research is needed to inform norms-based intervention strategies. Prospective studies with diverse sexual and gender minority samples should seek to address critical gaps in the literature pertaining to injunctive ATOD norms, social identification with meaningful reference groups, environmental influences on perceptions of substance use norms, and relationships between stigma experiences, perceptions of substance use norms, and substance use behaviors.

Public Significance Statement.

Sexual minorities disproportionately engage in substance use and experience dependence; however, targeted evidence-based substance use interventions for these populations remain alarmingly scarce. This scoping review seeks to encourage and inform intervention development by characterizing and synthesizing findings from social norms-focused studies over the past 20 years, highlighting social norms-based intervention and strategies that may have utility in reducing substance use in sexual minority populations, and identifying critical gaps in the literature to be addressed by future research.

Appendix A. PubMed Search Terms and Syntax

Sexual Orientation/Sexual Minority Related Terms	Sexual Orientation[TIAB] OR Sexual minority[TIAB] OR Sexual minorities[TIAB] OR Same-sex attraction [TIAB] OR Same-sex attracted [TIAB] Same-sex behavior[TIAB] OR Homosexual [TIAB] OR Homosexuality [TIAB] OR Queer[TIAB] OR Lesbian [TIAB] OR Lesbianism [TIAB] OR Women who have sex with women [TIAB] OR WSW [TIAB] OR Bisexual [TIAB] OR Bisexuality [TIAB] OR Women who have sex with women and men [TIAB] OR WSWM [TIAB] OR WSMW [TIAB] Men who have sex with men and women [TIAB] OR MSWM [TIAB] OR MSMW [TIAB] OR Pansexual[TIAB] OR Gay [TIAB] OR men who have sex with men[TIAB] OR MSM[TIAB] OR LGB[TIAB] OR LGBT[TIAB] OR LGBTQ[TIAB] OR GLBT[TIAB] OR GLB[TIAB] OR GLBTQ[TIAB]
*AND
Norm-Related Terms	Norm[TIAB] OR Norms[TIAB] OR Normative[TIAB] OR Normalization[TIAB] OR Normalisation[TIAB] OR Culture[TIAB] OR Social factors[TIAB] OR Social influences[TIAB] OR Community factors[TIAB] OR Community influences[TIAB] OR Cultural factors[TIAB] OR Conformity[TIAB] OR Socialization[TIAB]
*AND
Substance-Related Terms	Substance[TIAB] OR Substances[TIAB] OR Substance use[TIAB] OR Substance misuse[TIAB] OR Substance Abuse[TIAB] OR Alcohol[TIAB] OR Alcoholism [TIAB] OR Alcohol Use[TIAB] OR Alcohol Abuse[TIAB] OR Alcohol misuse[TIAB] OR Drinking[TIAB] OR Hazardous drinking[TIAB] OR Heavy episodic drinking[TIAB] OR Binge drink[TIAB] OR heavy drinking[TIAB] OR drunk[TIAB] OR intoxicated[TIAB] OR intoxication[TIAB] OR Ethanol[TIAB] OR Drug use[TIAB] OR Drug abuse[TIAB] OR Drug addiction[TIAB] OR Club drugs[TIAB] OR Smoke[TIAB] OR Smoking[TIAB] OR Tobacco[TIAB] OR Nicotine[TIAB] OR Cigarette[TIAB] OR e-cigarette[TIAB] OR Analgesic[TIAB] OR Depressant[TIAB] OR Stimulant[TIAB] OR Hallucinogen[TIAB] OR Narcotic[TIAB] OR Heroin[TIAB] OR Morphine[TIAB] OR Fentanyl[TIAB] OR Codeine[TIAB] OR Barbiturate[TIAB] OR Tranquilizer[TIAB] OR Cocaine[TIAB] OR Crack[TIAB] OR Methamphetamine[TIAB] OR Amphetamine [TIAB] OR Meth[TIAB] OR Ecstasy[TIAB] OR MDMA[TIAB] OR LSD[TIAB] OR Acid[TIAB] OR Marijuana Ketamine Special K Phencyclidine PCP Marijuana Cannabis Hashish Poppers GHB Prescription drugs Recreational drugs Addiction Dependence Adderall Ritalin Peyote Mescaline Psilocybin Magic mushrooms Opium Opioids Gamma- hydroxybutyric acid 4-hydroxybutanoic acid Bath salts